Towards a Deep Reinforcement Learning Model of Master Bay Stowage Planning

Jaike van Twiller; Djordje Grbic; Rune Møller Jensen

doi:10.1007/978-3-031-43612-3_6

Towards a Deep Reinforcement Learning Model of Master Bay Stowage Planning

Machine Learning

Research output: Conference Article in Proceeding or Book/Report chapter › Article in proceedings › Research › peer-review

Abstract

Major liner shipping companies aim to solve the stowage planning problem by optimally allocating containers to vessel locations during a multi-port voyage. Due to a large variety of combinatorial aspects, a scalable algorithm to solve a representative problem is yet to be found. This paper will show that deep reinforcement learning can optimize a non-trivial master bay planning problem. Our experiments show that proximal policy optimization efficiently finds reasonable solutions, serving as preliminary evidence of the potential value of deep reinforcement learning in stowage planning. In future work, we will extend our architecture to address a full-featured master bay planning problem.

Original language	English
Title of host publication	ICCL 2023: Computational Logistics
Editors	J.R. Daduna, G. Liedtke, X. Shi, S. Voß
Number of pages	17
Volume	14239
Place of Publication	Berlin
Publisher	Springer
Publication date	7 Sept 2023
Pages	105-121
Article number	6
Chapter	Maritime Shipping
ISBN (Print)	978-3-031-43611-6
ISBN (Electronic)	978-3-031-43612-3
DOIs	https://doi.org/10.1007/978-3-031-43612-3_6
Publication status	Published - 7 Sept 2023
Event	International Conference on Computational Logistics - Berlin, Germany Duration: 6 Sept 2023 → 8 Sept 2023 Conference number: 14 https://www.iccl2023.uni-hamburg.de/en/program.html

Conference

Conference	International Conference on Computational Logistics
Number	14
Country/Territory	Germany
City	Berlin
Period	06/09/2023 → 08/09/2023
Internet address	https://www.iccl2023.uni-hamburg.de/en/program.html

Series	Lecture Notes in Computer Science
Volume	14239
ISSN	0302-9743

Keywords

Maritime logistics
Liner shipping
Stowage planning
Deep reinforcement learning
Markov decision processes

Access to Document

10.1007/978-3-031-43612-3_6

978-3-031-43612-3_6Final published version, 1.52 MB

Cite this

@inproceedings{3f8da0cf0e964f0da4282e9d50067227,

title = "Towards a Deep Reinforcement Learning Model of Master Bay Stowage Planning",

abstract = "Major liner shipping companies aim to solve the stowage planning problem by optimally allocating containers to vessel locations during a multi-port voyage. Due to a large variety of combinatorial aspects, a scalable algorithm to solve a representative problem is yet to be found. This paper will show that deep reinforcement learning can optimize a non-trivial master bay planning problem. Our experiments show that proximal policy optimization efficiently finds reasonable solutions, serving as preliminary evidence of the potential value of deep reinforcement learning in stowage planning. In future work, we will extend our architecture to address a full-featured master bay planning problem.",

keywords = "Maritime logistics, Liner shipping, Stowage planning, Deep reinforcement learning, Markov decision processes, Maritime logistics, Liner shipping, Stowage planning, Deep reinforcement learning, Markov decision processes",

author = "\{van Twiller\}, Jaike and Djordje Grbic and Jensen, \{Rune M{\o}ller\}",

year = "2023",

month = sep,

day = "7",

doi = "10.1007/978-3-031-43612-3\_6",

language = "English",

isbn = "978-3-031-43611-6",

volume = "14239",

series = "Lecture Notes in Computer Science",

publisher = "Springer",

pages = "105--121",

editor = "J.R. Daduna and G. Liedtke and X. Shi and S. Vo{\ss}",

booktitle = "ICCL 2023: Computational Logistics",

address = "Germany",

note = "International Conference on Computational Logistics, ICCL ; Conference date: 06-09-2023 Through 08-09-2023",

url = "https://www.iccl2023.uni-hamburg.de/en/program.html",

}

van Twiller, J , Grbic, D & Jensen, RM 2023, Towards a Deep Reinforcement Learning Model of Master Bay Stowage Planning. in JR Daduna, G Liedtke, X Shi & S Voß (eds), ICCL 2023: Computational Logistics. vol. 14239, 6, Springer, Berlin, Lecture Notes in Computer Science, vol. 14239, pp. 105-121, International Conference on Computational Logistics, Berlin, Berlin, Germany, 06/09/2023. https://doi.org/10.1007/978-3-031-43612-3_6

Towards a Deep Reinforcement Learning Model of Master Bay Stowage Planning. / van Twiller, Jaike ; Grbic, Djordje ; Jensen, Rune Møller.
ICCL 2023: Computational Logistics. ed. / J.R. Daduna; G. Liedtke; X. Shi; S. Voß. Vol. 14239 Berlin: Springer, 2023. p. 105-121 6 (Lecture Notes in Computer Science, Vol. 14239).

Research output: Conference Article in Proceeding or Book/Report chapter › Article in proceedings › Research › peer-review

TY - GEN

T1 - Towards a Deep Reinforcement Learning Model of Master Bay Stowage Planning

AU - van Twiller, Jaike

AU - Grbic, Djordje

AU - Jensen, Rune Møller

N1 - Conference code: 14

PY - 2023/9/7

Y1 - 2023/9/7

N2 - Major liner shipping companies aim to solve the stowage planning problem by optimally allocating containers to vessel locations during a multi-port voyage. Due to a large variety of combinatorial aspects, a scalable algorithm to solve a representative problem is yet to be found. This paper will show that deep reinforcement learning can optimize a non-trivial master bay planning problem. Our experiments show that proximal policy optimization efficiently finds reasonable solutions, serving as preliminary evidence of the potential value of deep reinforcement learning in stowage planning. In future work, we will extend our architecture to address a full-featured master bay planning problem.

AB - Major liner shipping companies aim to solve the stowage planning problem by optimally allocating containers to vessel locations during a multi-port voyage. Due to a large variety of combinatorial aspects, a scalable algorithm to solve a representative problem is yet to be found. This paper will show that deep reinforcement learning can optimize a non-trivial master bay planning problem. Our experiments show that proximal policy optimization efficiently finds reasonable solutions, serving as preliminary evidence of the potential value of deep reinforcement learning in stowage planning. In future work, we will extend our architecture to address a full-featured master bay planning problem.

KW - Maritime logistics

KW - Liner shipping

KW - Stowage planning

KW - Deep reinforcement learning

KW - Markov decision processes

KW - Maritime logistics

KW - Liner shipping

KW - Stowage planning

KW - Deep reinforcement learning

KW - Markov decision processes

U2 - 10.1007/978-3-031-43612-3_6

DO - 10.1007/978-3-031-43612-3_6

M3 - Article in proceedings

SN - 978-3-031-43611-6

VL - 14239

T3 - Lecture Notes in Computer Science

SP - 105

EP - 121

BT - ICCL 2023: Computational Logistics

A2 - Daduna, J.R.

A2 - Liedtke, G.

A2 - Shi, X.

A2 - Voß, S.

PB - Springer

CY - Berlin

T2 - International Conference on Computational Logistics

Y2 - 6 September 2023 through 8 September 2023

ER -

Towards a Deep Reinforcement Learning Model of Master Bay Stowage Planning

Abstract

Conference

Keywords

Access to Document

Fingerprint

Cite this